mds: check metadata pool not cluster is full #19602
Merged
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
batrick
force-pushed
the
i22483
branch
3 times, most recently
from
e33134d
to
39d07b3
Compare
batrick
changed the title
liewegas
reviewed
Dec 19, 2017
| @@ -1720,7 +1720,9 @@ void Server::handle_osd_map() | |||
| * using osdmap_full_flag(), because we want to know "is the flag set" | |||
| * rather than "does the flag apply to us?" */ | |||
| mds->objecter->with_osdmap([this](const OSDMap& o) { | |||
| is_full = o.test_flag(CEPH_OSDMAP_FULL); | |||
There was a problem hiding this comment.
auto pi = o.get_pg_pool(mds->mdsmap->get_metadata_pool());
is_full = pi && pi->has_flag(pg_pool_t::FLAG_FULL);
There was a problem hiding this comment.
I was looking for that and couldn't find the right thing. Thanks!
liewegas
reviewed
Dec 19, 2017
src/mds/Server.cc
Outdated
| @@ -1720,7 +1720,9 @@ void Server::handle_osd_map() | |||
| * using osdmap_full_flag(), because we want to know "is the flag set" | |||
| * rather than "does the flag apply to us?" */ | |||
| mds->objecter->with_osdmap([this](const OSDMap& o) { | |||
| is_full = o.test_flag(CEPH_OSDMAP_FULL); | |||
| std::set<int64_t> full, backfillfull, nearfull; | |||
| o.get_full_pools(mds->cct, &full, &backfillfull, &nearfull); | |||
There was a problem hiding this comment.
this is used by the mon to set/clear the pool flags (and is slow). we only need to watch the flags.
CEPH_OSDMAP_FULL flag was obsoleted by b4ca5ae. So, check if the metadata pool is full instead which is a decent proxy (as metadata operations can still proceed). However, the data pool may still be full which would result in some operations still not completing (like inode backtrace updates). Fixes: http://tracker.ceph.com/issues/22483 Signed-off-by: Patrick Donnelly <pdonnell@redhat.com>
liewegas
approved these changes
Dec 19, 2017
|
The data pool full check is still missing? |
batrick
added a commit
to batrick/ceph
that referenced
this pull request
Dec 21, 2017
* refs/pull/19602/head: mds: check metadata pool not cluster is full
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
CEPH_OSDMAP_FULL flag was obsoleted by
b4ca5ae. So, check if the metadata pool is
full instead which is a decent proxy (as metadata operations can still
proceed). However, the data pool may still be full which would result in some
operations still not completing (like inode backtrace updates).
Fixes: http://tracker.ceph.com/issues/22483
Signed-off-by: Patrick Donnelly pdonnell@redhat.com